Eweek magazine is the latest to jump on the “Green Issue” trend this month (enough already). Within it are the numbers I have been looking for to justify the very ungreen practice I have of never shutting off my computers. As a CTO and software guy, I am at my computer throughout the day and night, and I can never stand to wait through the 2-3 minute boot time so I can do a quick check of whatever information I need at the moment. It seems that many corporations are frowning at this practice, installing software on their computers to shut them off each night, or after certain periods of inactivity. Eweek says “…the savings can be significant…you can save anywhere from $25 to $75 per PC per year by using the power management features on your PC.”

Let’s run the numbers, shall we? We’ll take the high side, at $75 and compare that to the time wasted waiting for the PC to boot. Best case, we’ll say it takes 2 minutes to be completely up and running, with all required software up and running also. At 50 weeks per year (2 off for vacation), that $75 savings is a break even point for an employee that gets paid … wait for it … $9 an hour; that is, if the employee makes more than $9 an hour, the two minutes they waste each day waiting for the PC to boot costs you more than $75 a year. It doesn’t take a rocket scientist to see that turning off the computer makes no business sense. Does it “save the earth’s resources” by using less electricity? Of course. Does it make any kind of economic sense? Of course not. However, idling the processor when not in use makes sense, but shutting down the machine entirely does not.

A mentor engineer of mine at my first job showed me how many of these “save a little here, save a little there” types of decisions make no sense, especially in high-tech companies. I remember his point about getting the engineering managers to not be so cheap, and buy the engineers a new computer every once in a while. We wrote our code in C, so we were constantly running compile cycles. He demonstrated that if a faster computer, over all your compile cycles that day, saved you just 10 minutes a day, for a $75k engineer, that works out to be a $1,562.50 savings for the entire year. Hence, a brand new $1200 computer once a year could be a savings to the company. Imagine that.

It pays to run the numbers on all the silly cost savings corporations try to push down the ladder. They’re better off spending some money and getting down to business.